National Audit Office chief Amyas Morse sums up the contents of the NAO's report on the work programme: "The Department for Work and Pensions has made a significant effort to learn the lessons of previous welfare to work programmes.

"It is too early to judge the success of the work programme, which will depend on whether the department can get more people into work than previous programmes.

"The department has set providers stretching performance targets and it needs to ensure that they do not cut corners to stay in profit, such as targeting easy to reach people, reducing service levels or treating sub-contractors unfairly."

The report concludes that the government overegged the impact of its work programme:

"The work programme's feasibility is underpinned by assumptions about likely performance but there is a significant risk that they are over-optimistic.

"The department expects that 36 per cent of people referred to providers will be placed into jobs for which providers will be paid.

"Our analysis of likely performance of the largest group of participants in the work programme (and one of the easiest to help into work) is that 26 per cent will get such jobs compared to the department's estimate of 40 per cent for that group.

"The department's estimate of performance and of the non-intervention rate – the percentage of participants that would have got work without the help of the work programme – was a major factor in determining the prices and performance incentives it set.

"If the performance estimates are too low then there is a risk that providers will make excessive profits.

"If these estimates are too high, prices will have been set too low and providers will find it difficult to meet minimum performance targets and struggle financially."